Pathways to sex

I was talking about sex and nothing but sex all last week in genetics, which is far less titillating than it sounds. My focus was entirely on operational genetics, that is, how autosomal inheritance vs inheritance of factors on sex chromosomes differ, and I only hinted at how sex is not inherited as a simple mendelian trait, as we're always tempted to assume, but is actually the product of a whole elaborate chain of epistatic interactions. I'm always tempted in this class to go full-blown rabid developmental geneticist on them and do nothing but talk about interactions between genes, but I manage to restrain myself every time -- we have a curriculum and a focus for this course, and it's basic transmission genetics, and I struggle to get general concepts across before indulgence in my specific interests. Stick to the lesson plan! Try not to break everyone's brain yet!

But a fellow can dream, right?

Anyway, before paring everything down to the reasonable content I can give in a third year course, I brush up on the literature and take notes and track down background and details that I won't actually dump on the students (fellow professors know this phenomenon: you have to work to keep well ahead of the students, because they really don't need to start thinking they're smarter than you are). But I can dump my notes on you. You don't have to take a test on it and get a good grade, and you won't pester me about whether this will actually be on the test, and you won't start crying if I overwhelm you with really cool stuff. (If any of my students run across this, no, the content of this article will not be on any test. Don't panic. Go to grad school where this will all be much more relevant.)

Onward. Here's my abbreviated summary of the epistatic interactions in making boys and girls.

The earliest step in gonad development is the formation of the urogenital ridge from intermediate mesoderm, a thickening on the outside of the mesonephros (early kidney), under the influence of transcription factors Emx2, Wt1 (Wilms tumor 1), Lhx9, and Sf1 (steroidogenic factor 1). Even in the earliest stages, multiple genes interact to generate the tissue! The urogenital ridge is going to form only the somatic tissue of the gonad; the actual germ cells (the cells that will form the gametes, sperm and ova) arise much, much earlier, in the epiblast of the embryo at a primitive streak stage, and then migrate through the mesenteries of the gut to populate the urogenital ridge independently, shortly after it forms.

At this point, this organ is called the bipotential gonad -- it is identical in males and females. Two genes, Fgf9 and Wnt4, teeter in a balanced antagonistic relationship -- Wnt4 suppresses Fgf9, and Fgf9 suppresses Wnt4 -- in the bipotential gonad, and anything that might tip the balance between them will trigger development of one sex or the other. A mutation that breaks Fgf9, for instance, gives Wnt4 an edge, and the gonad will develop into an ovary; a mutation that breaks Wnt4 will let Fgf9 dominate the relationship, and the gonad will develop into a testis (with a note of caution: the changes will initiate differentiation into one gonad or the other, but there are other steps downstream that can also vary). These two molecules may be the universal regulators of the sex of the gonad in animals: fruit flies also use Fgf and Wnt genes to regulate development of their gonads.

But the key to the genetic symmetry-breaking of selecting Fgf9 and Wnt4 varies greatly in animals. Some use incubation temperature to bias expression one way or the other; birds have a poorly understood set of factors that may require heterodimerization between two different proteins produced on the Z and W chromosomes to induce ovaries; mammals have a unique gene, Sry, not found in other vertebrates, that is located on the Y chromosome and tilts the balance towards testis differentiation.

Sry may be unique to mammals, but it didn't come out of nowhere. Sry contains a motif called the HMG (high mobility group) box, which is a conserved DNA binding domain. There are approximately 20 proteins related to Sry in humans, all given the name SOX, for SRY-related HMG box (I know, molecular biologists seem to be really reaching for acronyms nowadays). SOX genes are found in all eukaryotes, and seem to play important roles in cell and organ differentiation in insects, nematodes, and vertebrates. Sry is simply the member of the family that has been tagged to regulate gonad development in mammals.

If a copy of Sry is present in the organism, which is usually only the case in XY or male mammals, expression of the gene produces a DNA binding protein that has one primary target: a gene called SOX9 (they're cousins!). In mice, Sry is switched on only transiently, long enough to activate SOX9, which then acts as a transcription factor for itself, maintaining expression of SOX9 for the life of the gonad. Humans keep Sry turned on permanently as well, but there's no evidence yet that it actually does anything important after activating SOX9; it may be that human males neglect to hit the off switch.

SOX9 binds to a number of genes, among them, Fgf9. Remember Fgf9? The masculinizing factor in antagonism to the feminizing factor Wnt4? This tips the teeter-totter to favor expression of Fgf9 over Wnt4, leading to the differentiation of a testis from the bipotential gonad.

So far, then, we've got a nice little Rube Goldberg machine and an epistatic pathway. Sf1/Wt1 and other early genes induce the formation of a urogenital ridge and an ambiguous gonad; Sry upregulates Sox9 which upregulates Fgf9 which suppresses Wnt4, turning off the ovarian pathway and turning on the testis pathway.

But wait, we're not done! Sry/SOX9 are expressed specifically in a subset of cells of the male gonad, the prospective Sertoli cells. If you recall your reproductive physiology, Sertoli cells are a kind of 'nurse' cell of the testis; they're responsible for nourishing developing sperm cells. They also have signaling functions. The Sertoli cells produce AMH, or anti-Müllerian Hormone, which is responsible for causing the female ducts of the reproductive system to degenerate in males (if you don't remember the difference between Müllerian and Wolffian and that array of tubes that get selected for survival in the different sexes, here's a refresher). Defects in the AMH system lead to persistent female ducts: you get males with partial ovaries and undescended testicles. So just having the Sry chain is not enough, there are downstream genes that have to dismantle incipient female structures and promote mature properties of the gonad.

As the gonad differentiates, it also induces another set of cells, the embryonic Leydig cells. We have to distinguish embryonic Leydig cells, because they represent another transient population that will do their job in the embryo, then gradually die off to be replaced by a new population of adult Leydig cells at puberty. The primary function of Leydig cells is the production of testosterone and other androgens. The embryo gets a brief dose of testosterone early that initiates masculinization of various tissues, which then fades (fortunately; no beards and pubic hair for baby boys) to resurge in adolescence, triggering development of secondary sexual characteristics. Embryonic testosterone is the signal that maintains the Wolffian duct system. No testosterone, and the Wolffian ducts degenerate.

Just to complicate matters, while testosterone is the signal that regulates the male ducts, testosterone must be converted to dihydrotestosterone (DHT), the signal that regulates development of the external genitalia. Defects in the enzyme responsible for this conversion can lead to individuals with male internal plumbing, including testes, but female external genitalia. Sex isn't all or nothing, but a whole series of switches!

By now, if you're paying attention, you may have noticed a decidedly male bias in this description. I've been talking about a bipotential gonad that is flipped into a male mode by the presence of a single switch, and sometimes, especially in the older literature, you'll find that development of the female gonad is treated as the default: ovaries are what you get if you lack the special magical trigger of the Y chromosome. This is not correct. The ovaries are also the product of an elaborate series of molecular decisions; I think it's just that they Y chromosome and the Sry gene just provided a convenient genetic handle to break into the system, and really, scientists usually favor the easy tool to get in.

One key difference between the testis and ovary is the inclusion of germ cells. The testis simply doesn't care; if the germ line, the precursors to sperm, is not present, the male gonad goes ahead and builds cords of Sertoli cells with Leydig cells differentiating in the interstitial space, makes the whole dang structure of the testicle, pumping out testosterone as if all is well, but contains no cells to make sperm — so it's reproductively useless, but hormonally and physiologically active. The ovary is different. If no germ line populates it, the ovarian follicle cells (the homolog to the Sertoli cells) do not differentiate. If germ cells are lost from the tissue only later, the follicles degenerate.

Ovaries require a signal from the germ line to develop normally. One element of that signal seems to be factors associated with cells in meiosis. The female germ line cells are on a very strict meiotic clock, beginning the divisions to produce haploid egg cells in the embryo, even as they populate the gonad. These oocytes produce a signal, Figα (factor in germ line a) that recruits ovarian cells to produce follicles. The male gonad has to actively repress meiosis in the embryonic germ line to inhibit this signaling; male germ cells are restricted to only mitotic divisions until puberty.

Even before Figα signaling becomes important, there are other factors uniquely expressed in the prospective ovary that shape its development. In particular, Wnt4 induces the expression of another gene, Foxl2, that is critical for formation of the ovarian follicle. The pathways involved in ovarian development are not as well understood as those in testis development, but it's quite clear that there is a chain of specific genetic/molecular interactions involved in the generation of both organs.

Wait, you say, you need a diagram! You can't grasp all this without an illustration! Here's a nice one: I particularly like that cauliflower-shaped explosion of looping arrows early in the testis pathway.

The molecular and genetic events in mammalian sex determination. The bipotential genital ridge is established by genes including Sf1 and Wt1, the early expression of which might also initiate that of Sox9 in both sexes. b-catenin can begin to accumulate as a response to Rspo1–Wnt4 signaling at this stage. In XX supporting cell precursors, b-catenin levels could accumulate sufficiently to repress SOX9 activity, either through direct protein interactions leading to mutual destruction, as seen during cartilage development, or by a direct effect on Sox9 transcription. However, in XY supporting cell precursors, increasing levels of SF1 activate Sry expression and then SRY, together with SF1, boosts Sox9 expression. Once SOX9 levels reach a critical threshold, several positive regulatory loops are initiated, including autoregulation of its own expression and formation of feed-forward loops via FGF9 or PGD2 signaling. If SRY activity is weak, low or late, it fails to boost Sox9 expression before b-catenin levels accumulate sufficiently to shut it down. At later stages, FOXL2 increases, which might help, perhaps in concert with ERs, to maintain granulosa (follicle) cell differentiation by repressing Sox9 expression. In the testis, SOX9 promotes the testis pathway, including Amh activation, and it also probably represses ovarian genes, including Wnt4 and Foxl2. However, any mechanism that increases Sox9 expression sufficiently will trigger Sertoli cell development, even in the absence of SRY.

So that's what I didn't tell my genetics students this time around. Maybe I'll work it into my developmental biology course, instead.

Kim Y, Kobayashi A, Sekido R, DiNapoli L, Brennan J, Chaboissier MC, Poulat F, Behringer RR, Lovell-Badge R, Capel B. (2006) Fgf9 and Wnt4 act as antagonistic signals to regulate mammalian sex determination. PLoS Biol 4(6):e187

Ross AJ, Capel B. (2005) Signaling at the crossroads of gonad development. Trends Endocrinol Metab. 16(1):19-25.

Sekido R, Lovell-Badge R (2009) Sex determination and SRY: down to a wink and a nudge? Trends Genet. 25(1):19-29.

Sim H, Argentaro A, Harley VR (2008) Boys, girls and shuttling of SRY and SOX9. Trends Endocrinol Metab. 19(6):213-22.

Yao H H-C (2005) The pathway to femaleness: current knowledge on embryonic
development of the ovary. Molecular and Cellular Endocrinology 230:87–93.